feat(ci): add PyPI publish workflow

.github/workflows/publish.yml

@@ -0,0 +1,32 @@
+name: Publish to PyPI
+
+on:
+  push:
+    tags:
+      - "v*"
+
+jobs:
+  build-and-publish:
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      id-token: write
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+
+      - name: Install build tools
+        run: pip install build
+
+      - name: Build wheel and sdist
+        run: python -m build
+
+      - name: Publish to PyPI
+        uses: pypa/gh-action-pypi-publish@release/v1
+        with:
+          password: ${{ secrets.PYPI_API_TOKEN }}

PRIVACY.md

@@ -0,0 +1,58 @@
+# Privacy Policy
+
+**Effective date:** 2026-04-21
+
+This Privacy Policy describes how the **iotcli** open-source project ("we", "us", or "our") handles information.
+
+## 1. Overview
+
+iotcli is a local command-line tool for controlling IoT devices. It does not operate any central service, does not collect telemetry, and does not transmit your personal data to project maintainers.
+
+## 2. Information Stored Locally
+
+All data created by iotcli is stored locally on your machine in `~/.iotcli/`:
+
+| File / Directory | Contents |
+|------------------|----------|
+| `devices.yaml`   | Device names, protocols, IP addresses, and non-sensitive settings |
+| `credentials/*.enc` | Fernet-encrypted secrets (tokens, local keys, passwords) |
+| `.key`           | Local-only Fernet encryption key (permissions `0600`) |
+| `skills/`        | Generated AI-agent skill files based on your device configuration |
+
+No project maintainer, server, or third party (other than the device vendor you are intentionally communicating with) has access to these files.
+
+## 3. Third-Party Services
+
+When you use iotcli to communicate with a device, data is transmitted directly between your machine and the device or its vendor cloud (e.g., Tuya cloud, LG ThinQ). This communication is:
+
+- **Initiated by you** via explicit CLI commands.
+- ** Governed by the device vendor's privacy policy**, not this one.
+
+We do not act as an intermediary, proxy, or data processor for these communications.
+
+## 4. No Telemetry or Analytics
+
+iotcli does not:
+
+- Collect usage statistics or crash reports.
+- Use tracking cookies or fingerprinting.
+- Send data to external analytics services.
+- Display advertisements.
+
+## 5. Data Security
+
+- Credentials are encrypted at rest using Fernet (symmetric AES-128 in CBC mode with HMAC).
+- The encryption key is stored with owner-only read permissions (`0600`).
+- We recommend you keep your operating system and Python environment up to date.
+
+## 6. Open Source
+
+The full source code is available at https://github.com/fomyio/iotcli. You can inspect exactly what the tool does and how it handles data.
+
+## 7. Changes to This Policy
+
+We may update this policy to reflect changes in the software. The latest version will always be available in the repository.
+
+## 8. Contact
+
+For privacy questions, open an issue at https://github.com/fomyio/iotcli/issues.

README.md

@@ -1,33 +1,56 @@
 # iotcli
 
-**Give your AI agent hands — control any smart home device from the terminal.**
-
-One CLI to rule them all. Discover, configure, and control IoT devices across protocols. Built for AI agents, loved by humans.
-
-```text
-$ iotcli control status lg-ac
-╭─── lg-ac — online ───╮
-│   power: POWER_OFF   │
-│   mode: HEAT         │
-│   fan_speed: HIGH    │
-│   current_temp: 20.5 │
-│   target_temp: 25    │
-╰──────────────────────╯
-```
+<p align="center">
+  <b>Give your AI agent hands.</b><br>
+  One CLI to rule them all. Discover, configure, and control IoT devices across protocols.<br>
+  Built for AI agents, loved by humans.
+</p>
+
+<p align="center">
+  <a href="https://pypi.org/project/iotcli/"><img src="https://img.shields.io/pypi/v/iotcli?style=flat-square&color=blue" alt="PyPI version"></a>
+  <a href="https://pypi.org/project/iotcli/"><img src="https://img.shields.io/pypi/pyversions/iotcli?style=flat-square" alt="Python versions"></a>
+  <a href="https://github.com/fomyio/iotcli/actions"><img src="https://img.shields.io/github/actions/workflow/status/fomyio/iotcli/ci.yml?style=flat-square&logo=github" alt="CI"></a>
+  <a href="https://codecov.io/gh/fomyio/iotcli"><img src="https://img.shields.io/codecov/c/github/fomyio/iotcli?style=flat-square&logo=codecov" alt="Coverage"></a>
+  <a href="LICENSE"><img src="https://img.shields.io/badge/license-MIT-green?style=flat-square" alt="License"></a>
+  <a href="PRIVACY.md"><img src="https://img.shields.io/badge/privacy-local%20only-success?style=flat-square" alt="Privacy"></a>
+</p>
+
+<p align="center">
+  <a href="#install">Install</a> &bull;
+  <a href="#quick-start">Quick Start</a> &bull;
+  <a href="#ai-agent-integration">AI Integration</a> &bull;
+  <a href="#supported-protocols">Protocols</a> &bull;
+  <a href="#architecture">Architecture</a>
+</p>
+
+---
+
+## What is iotcli?
+
+iotcli is a universal IoT device control CLI. It unifies smart-home devices behind a single, predictable interface — so you (or your AI agent) can control lights, AC units, feeders, and sensors without memorizing protocol quirks.
+
+**Key principles:**
+- **One command shape** for every device — `on`, `off`, `status`, `set`
+- **Protocol-agnostic** — Tuya, miIO, MQTT, HTTP, LG ThinQ, and more
+- **AI-native** — JSON mode, skill files, and an MCP server for Claude Desktop / Cursor / Zed
+- **Privacy-first** — everything local, credentials encrypted, no telemetry
 
 ## Install
 
 ```bash
-# Global install (recommended — no venv needed, no package conflicts)
-pipx install git+https://github.com/joeVenner/CLI-IoT.git
+# Global install via pipx (recommended — no venv needed)
+pipx install git+https://github.com/fomyio/iotcli.git
+
+# Or with MCP support for Claude Desktop / Cursor
+pipx install git+https://github.com/fomyio/iotcli.git[mcp]
 
-# Or from a local clone
-git clone https://github.com/joeVenner/CLI-IoT.git
-cd CLI-IoT
-pipx install .
+# From source
+git clone https://github.com/fomyio/iotcli.git
+cd iotcli
+pip install -e ".[dev]"
 ```
 
-After install, `iotcli` is available system-wide. No activation, no venv.
+> Requires Python 3.10+
 
 ## Quick Start
 
@@ -36,9 +59,9 @@ After install, `iotcli` is available system-wide. No activation, no venv.
 iotcli setup
 
 # Discover devices on your network
-iotcli discover
+iotcli discover --network 192.168.1.0/24
 
-# Add a device non-interactively
+# Add a device
 iotcli add --name living-room-light --protocol miio \
     --ip 192.168.1.100 --token <32chars>
 
@@ -53,24 +76,48 @@ iotcli --json control status living-room-light
 iotcli --json status-all
 ```
 
+## Demo
+
+```text
+$ iotcli list
+╭─────────── Devices ───────────╮
+│ Name              Protocol  IP            │
+│ living-room-light miio      192.168.1.100 │
+│ feeder            tuya      192.168.1.4   │
+│ lg-ac             lgac      (cloud)       │
+╰───────────────────────────────╯
+
+$ iotcli control status lg-ac
+╭─── lg-ac — online ───╮
+│   power: POWER_OFF   │
+│   mode: HEAT         │
+│   fan_speed: HIGH    │
+│   current_temp: 20.5 │
+│   target_temp: 25    │
+╰──────────────────────╯
+
+$ iotcli --json control on feeder
+{"success": true, "device": "feeder", "action": "on"}
+```
+
 ## Supported Protocols
 
-| Protocol | Devices                          | Connection    |
-| -------- | -------------------------------- | ------------- |
-| `miio`   | Xiaomi / Yeelight                | Local (UDP)   |
-| `tuya`   | Tuya-based (lights, plugs, etc.) | Local (TCP)   |
-| `mqtt`   | Zigbee / Aqara via MQTT broker   | Local (TCP)   |
-| `http`   | ESPHome / Tasmota                | Local (HTTP)  |
-| `lgac`   | LG Air Conditioner (ThinQ)       | Cloud (HTTPS) |
+| Protocol | Devices | Connection |
+|----------|---------|------------|
+| `miio` | Xiaomi / Yeelight | Local (UDP) |
+| `tuya` | Tuya-based (lights, plugs, etc.) | Local (TCP) |
+| `mqtt` | Zigbee / Aqara via MQTT broker | Local (TCP) |
+| `http` | ESPHome / Tasmota | Local (HTTP) |
+| `lgac` | LG Air Conditioner (ThinQ) | Cloud (HTTPS) |
 
-### Tuya Device Profiles
+### Tuya Profiles
 
-| Profile      | Devices               | Special Actions                        |
-| ------------ | --------------------- | -------------------------------------- |
-| `generic`    | Any Tuya device       | power on/off                           |
-| `light`      | Smart bulbs           | brightness, color_temperature, color   |
-| `switch`     | Smart plugs           | power, countdown                       |
-| `petfeeder`  | ROJECO / Tuya feeders | portions, quick_feed, slow_feed, light |
+| Profile | Devices | Special Actions |
+|---------|---------|-----------------|
+| `generic` | Any Tuya device | power on/off |
+| `light` | Smart bulbs | brightness, color_temperature, color |
+| `switch` | Smart plugs | power, countdown |
+| `petfeeder` | ROJECO / Tuya feeders | portions, quick_feed, slow_feed, light |
 
 ```bash
 iotcli add --name feeder --protocol tuya --profile petfeeder \
@@ -79,32 +126,35 @@ iotcli add --name feeder --protocol tuya --profile petfeeder \
 
 ## AI Agent Integration
 
-Every command supports `--json` for structured output. Feed your agent the generated skill files and it knows how to control your home.
+iotcli was built from the ground up for AI agents. Every command supports `--json` for structured, parseable output.
+
+### Skill Files
+
+Generate per-device skill files so your agent knows exactly what each device can do:
 
 ```bash
-# Generate skill files for all configured devices
 iotcli skills generate
-
-# Files created in ~/.iotcli/skills/:
-#   iotcli.skill.yaml     — global tool spec with all devices
-#   system_prompt.md      — ready-to-use agent system prompt
-#   <device>/SKILL.md     — per-device capability doc
-#   iotcli.tools.json     — OpenAI/Anthropic tool schema
 ```
 
+Files created in `~/.iotcli/skills/`:
+- `<device>/SKILL.md` — per-device capability doc (OpenClaw-compatible)
+- `iotcli.tools.json` — OpenAI/Anthropic tool schema
+- `iotcli.skill.yaml` — legacy global skill spec
+- `system_prompt.md` — ready-to-use agent system prompt
+
 ### MCP Server (Claude Desktop / Cursor / Zed)
 
-iotcli ships an MCP server so Claude can control your devices directly in conversation — no copy-pasting commands.
+Connect iotcli directly to your AI assistant via the Model Context Protocol:
 
 ```bash
 # Install with MCP support
 pip install iotcli[mcp]
 
-# Start the server (Claude Desktop launches this for you)
+# Start the server
 iotcli serve mcp
 ```
 
-Add to your Claude Desktop config (`claude_desktop_config.json`):
+Add to `claude_desktop_config.json`:
 
 ```json
 {
@@ -117,9 +167,9 @@ Add to your Claude Desktop config (`claude_desktop_config.json`):
 }
 ```
 
-Once connected, Claude can list devices, check status, turn them on/off, and set properties — all through structured tool calls with per-device enums and value validation.
+Once connected, Claude can list devices, check status, turn them on/off, and set properties — all through structured tool calls with per-device enums and value validation. No copy-pasting commands.
 
-### Agent Workflow (CLI)
+### Agent Workflow
 
 ```bash
 # 1. Discover what's available
@@ -133,6 +183,36 @@ iotcli --json control on "living-room-light"
 iotcli --json control set "living-room-light" "brightness=80"
 ```
 
+## Architecture
+
+```text
+src/iotcli/
+├── core/           # Device model, protocol registry, controller
+├── protocols/      # Self-registering protocol handlers
+├── config/         # YAML config + Fernet credential vault
+├── discovery/      # Async multi-protocol network scanner
+├── tui/            # Rich + InquirerPy interactive wizard
+├── cli/            # Click commands (discover, control, device, config, skills)
+├── skills/         # Jinja2-based AI agent skill generator
+└── mcp/            # MCP server for Claude Desktop / Cursor / Zed
+```
+
+**Design decisions:**
+- **Protocol registry** (`core/registry.py`): Decorator-based self-registration. No hardcoded mappings.
+- **Device model** (`core/device.py`): Dataclass with `slugify()` for normalized names.
+- **Credentials**: Fernet-encrypted vault, never in `devices.yaml`. Key stored at `0600`.
+- **Extensible**: Add a protocol by implementing `BaseProtocol` — the CLI, wizard, and skill generator pick it up automatically.
+
+## Security
+
+- Credentials are **never** stored in config files
+- Secrets encrypted with Fernet at `~/.iotcli/credentials/`
+- Encryption key has `0600` permissions (owner-only)
+- No telemetry, no analytics, no cloud dependency
+- Fully open source — inspect exactly what runs on your machine
+
+See [PRIVACY.md](PRIVACY.md) for details.
+
 ## Extending — Add a New Protocol
 
 ```python
@@ -158,12 +238,12 @@ class MyProtocol(BaseProtocol):
 
 Add one import in `protocols/__init__.py` — the CLI, wizard, and skill generator pick it up automatically.
 
-## Security
-
-- Credentials are **never** stored in config files
-- Secrets encrypted with Fernet at `~/.iotcli/credentials/`
-- Encryption key has `0600` permissions (owner-only)
-
 ## License
 
 MIT
+
+---
+
+<p align="center">
+  Made with passion for AI agents and smart homes.
+</p>

src/iotcli/cli/commands/control.py

@@ -38,7 +38,6 @@ def control(ctx, action, device_name, value):
     out = Output(ctx.obj["json_output"])
     cfg: ConfigManager = ctx.obj["config"]
 
-    device = cfg.get_device_or_none(device_name)
     device = cfg.get_device_or_none(device_name)
     if not device:
         out.error(f"Device not found: {device_name}")